Effective AF points
Why do AF points only work towards
the centre of the sensor in a DSLR and is this likely to change in the future?
However
a manually assignable AF point can be moved around
Autofocus systems work by detecting contrast
differences between adjacent pixels in the sensor. The higher the contrast
difference between the pixels, the more accurate the focusing will be.
All the AF points in a digital SLR should
work effectively, though the central point is always the most sensitive. The
number and distribution of AF points varies from model to model. If you have an
entry-level digital SLR then it may have a relatively low number of AF points
and those points will generally be towards the centre of the sensor. However, high-end
DSLRs have dozens of AF points (Canon's EOS 5D Mk lllin has 61!) covering most
of the sensor area so that subjects off-centre can still be picked-up by the AF
system. AF points are clustered more towards the centre of the image area
simply because that's where we tend to place our main subject. It's unlikely
you're going to put an important part of a shot at the very edges of the frame,
so there's no need to have AF points there. That said, as the number of AF
points in DSLRs increases, you may soon find that they do progress to cover the
entire image area.
I'm off travelling soon and intend to
upload images to my website, Flickr and Facebook as I go. If I photograph local
people, do I need to get them to sign a model release form?
AF
point display via the Quick Control menu in Live
View
For the uses you're talking about, no - you
don't need model release forms. There's no law against putting images of people
on websites and photo-sharing sites without their permission, providing you
don't use the images in such a way that could be construed as defamation of
character - for example, referring to someone as a drunk if you photograph them
having a drink. Even if you did, the chance of the person ever finding out and
taking action is very slim, but you should respect the people you encounter and
photograph on your travels.
If at a later date you decide to submit
some of your images to a photo library, you may be asked to provide model
releases for all people shots. If you don't have releases you need to declare
this. The library may then decide to either reject the images, or apply
restrictions to them. Again, there's no law against publishing photographs of
people without their permission, providing they are not depicted in a negative
way, so most libraries will accept non-released portraits for use in travel
guides, magazines, brochures etc. It's only really for advertising use that
model releases are required, because you can't use someone's face to endorse a
product or service without their permission - and a signed model release gives
you that permission.
Another factor to consider is the effect of
asking people you encounter to sign what is, in effect, a legally binding
contract moments after you've met them! It's quite a big ask and could spoil
what would otherwise have been a pleasant and casual experience!
Colour confusion
I am getting confused about what
would be the best set-up for me regarding colour space. My Nikon D90 gives me
the option of shooting in sRGB or Adobe RGB; I use Adobe RGB. For software, I
use Lightroom 3 which I set up to use ProPhoto RGB. I read that ProPhoto RGB
encompasses the Adobe RGB range and is slightly larger, so I figured it was
best to have the largest range possible. My monitor however is sRGB, so some
colours I would not be able to see. I figure that I would rather that than
limit myself in terms of colour. I haven't yet entered the world of printing,
but that is next on the agenda. Is it better to have the same colour space from
start to finish or is there some logic in my logic?!
Colour
confusion
We spoke to Simon Prais, technical director
at Color Confidence (www.colorconfidence.com) to gain an insight into the
different colour space options. The basic answer is that it is preferable not
to convert images from one RGB colour space to another during the image
editing, only when preparing them for print or web.
Converting from sRGB to the larger Adobe
RGB (or Pro Photo RGB) colour space will not improve your colours. The
conversion process will in fact negatively affect image quality. You're far
better off maximising the use of your current gamut by increasing colour
saturation.
Similarly, converting from a large colour
space like Adobe RGB to a smaller sRGB will reduce the colour gamut of your
image and should only be applied if the required output is sRGB, which it often
is for web images.
The more complex question is which colour
space should you being using? Working in Adobe RGB is recommended if you want
to benefit from saturated colours when it comes to print. Most professional
monitors now display the Adobe RGB gamut but even if they didn't, this colour
space can still be used so long as you're not too excessive with editing colour
saturation on an sRGB screen.
If you don't have a full understanding of
colour management and profile conversions, however, I think it's safer to stick
with sRGB. Working in sRGB has it advantages, for example, in situations where
your images lack strong saturated colours, such as in portrait photography. In
this instance, the range outside of sRGB isn't required and the smaller colour
space provides a smaller difference between subtle tones than the larger colour
spaces.
